6072 Background: It remains uncertain which chemotherapy sequence is more effective for locoregionally advanced nasopharyngeal carcinoma (NPC). We aimed to compare the efficacy and safety of induction-concurrent with concurrent-adjuvant chemotherapy in high-risk N2–3 NPC. Methods: We conducted an open-label, phase 3, single-centre, randomised controlled trial. Patients aged 18–65 years with stage T1-4N2-3M0 (AJCC 7th staging system) and pretreatment Epstein–Barr virus DNA level ≥1500 copies/ml were enrolled. Eligible patients were randomly assigned (1:1) to receive paclitaxel liposome (135 mg/m² intravenously on day 1), cisplatin (25 mg/m² intravenously on days 1-3), and fluorouracil (3.75g/m² continuous intravenous infusion for 120 h) induction chemotherapy once every 3 weeks, for 3 cycles, followed by concurrent cisplatin (100 mg/m² intravenously) on days 1, 22, and 43 of intensity-modulated radiotherapy or concurrent chemoradiotherapy followed by fluorouracil (4 g/m² in continuous intravenous infusion for 96 h) and cisplatin (80 mg/m² intravenously on day 1) once every 4 weeks, for 3 cycles. Randomisation was performed using computer-generated random number code with a block size of 6, stratified by nodal category (N2 or N3). The primary endpoint was 3-year progression-free survival (PFS) in the intention-to-treat population. This trial was registered with ClinicalTrials.gov (number NCT03306121). Results: Between 20 November 2017 and 19 March 2021, 162 patients were assigned to the induction-concurrent group and 162 to the concurrent-adjuvant group. Regarding the data cutoff (20 January 2024), the median follow-up period was 51.7 months. The 3-year PFS rates were 73.4% (95%CI 65.9%-79.5%) in the induction-concurrent group and 69.8% (95%CI 62.0%-76.2%) in the concurrent-adjuvant group (HR 0.85, [95%CI 0.58-1.26], P=0.43). Patients in the induction-concurrent group had significantly better 3-year distant metastasis-free survival rate at 81.4% (95% CI, 74.5%–86.6%) compared with those in the concurrent-adjuvant group at 71.0% (95% CI, 63.3%–77.3%) (HR, 0.64; 95% CI, 0.42–0.98; p=0.04). There were no differences on overall survival and locoregional failure-free survival between the two groups. The most common acute grade 3 + adverse events were leukopenia (53 [33.1%] of 160 in the induction-concurrent group vs. 47 [33.1%] of 142 in the concurrent-adjuvant group), neutropenia (51 [31.9%] vs. 32 [22.5%]), and mucositis (47 [29.4%] vs. 42 [29.6%]). The most common grade 3 + late adverse event was auditory or hearing loss (10 [6.3%] vs. 12 [8.5%]). Conclusions: Induction-concurrent chemotherapy does not significantly improve PFS compared with concurrent-adjuvant chemotherapy in high-risk N2–3 NPC. Long-term follow-up is required to determine long-term efficacy and toxicities. Clinical trial information: NCT03306121 .